mth/ebiten
1
0
Fork 0
mirror of https://github.com/hajimehoshi/ebiten.git synced 2024-12-25 11:18:54 +01:00
Code Issues Packages Projects Releases Wiki Activity

graphics: Optimization mainly for browsers

Browse Source
This commit is contained in:
Hajime Hoshi 2016-10-22 23:35:43 +09:00
parent af4130c0d6
commit 95f231d58a
1 changed files with 69 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

89
imageparts.go
View File

@ -85,11 +85,12 @@ type textureQuads struct {
func (t *textureQuads) vertices() []uint8 { func (t *textureQuads) vertices() []uint8 {
l := t.parts.Len() l := t.parts.Len()
vertices := make([]uint8, 0, l*32) vertices := make([]uint8, l*32)
p := t.parts p := t.parts
w, h := t.width, t.height w, h := t.width, t.height
width2p := int(graphics.NextPowerOf2Int32(int32(w))) width2p := int(graphics.NextPowerOf2Int32(int32(w)))
height2p := int(graphics.NextPowerOf2Int32(int32(h))) height2p := int(graphics.NextPowerOf2Int32(int32(h)))
n := 0
for i := 0; i < l; i++ { for i := 0; i < l; i++ {
dx0, dy0, dx1, dy1 := p.Dst(i) dx0, dy0, dx1, dy1 := p.Dst(i)
if dx0 == dx1 || dy0 == dy1 { if dx0 == dx1 || dy0 == dy1 {
@ -101,27 +102,75 @@ func (t *textureQuads) vertices() []uint8 {
continue continue
} }
u0, v0, u1, v1 := u(sx0, width2p), v(sy0, height2p), u(sx1, width2p), v(sy1, height2p) u0, v0, u1, v1 := u(sx0, width2p), v(sy0, height2p), u(sx1, width2p), v(sy1, height2p)
// Use direct assign here. `append` function might be slow on browsers.
if endian.IsLittle() { if endian.IsLittle() {
vertices = append(vertices, vertices[32*n] = uint8(x0)
uint8(x0), uint8(x0>>8), uint8(y0), uint8(y0>>8), vertices[32*n+1] = uint8(x0 >> 8)
uint8(u0), uint8(u0>>8), uint8(v0), uint8(v0>>8), vertices[32*n+2] = uint8(y0)
uint8(x1), uint8(x1>>8), uint8(y0), uint8(y0>>8), vertices[32*n+3] = uint8(y0 >> 8)
uint8(u1), uint8(u1>>8), uint8(v0), uint8(v0>>8), vertices[32*n+4] = uint8(u0)
uint8(x0), uint8(x0>>8), uint8(y1), uint8(y1>>8), vertices[32*n+5] = uint8(u0 >> 8)
uint8(u0), uint8(u0>>8), uint8(v1), uint8(v1>>8), vertices[32*n+6] = uint8(v0)
uint8(x1), uint8(x1>>8), uint8(y1), uint8(y1>>8), vertices[32*n+7] = uint8(v0 >> 8)
uint8(u1), uint8(u1>>8), uint8(v1), uint8(v1>>8)) vertices[32*n+8] = uint8(x1)
vertices[32*n+9] = uint8(x1 >> 8)
vertices[32*n+10] = uint8(y0)
vertices[32*n+11] = uint8(y0 >> 8)
vertices[32*n+12] = uint8(u1)
vertices[32*n+13] = uint8(u1 >> 8)
vertices[32*n+14] = uint8(v0)
vertices[32*n+15] = uint8(v0 >> 8)
vertices[32*n+16] = uint8(x0)
vertices[32*n+17] = uint8(x0 >> 8)
vertices[32*n+18] = uint8(y1)
vertices[32*n+19] = uint8(y1 >> 8)
vertices[32*n+20] = uint8(u0)
vertices[32*n+21] = uint8(u0 >> 8)
vertices[32*n+22] = uint8(v1)
vertices[32*n+23] = uint8(v1 >> 8)
vertices[32*n+24] = uint8(x1)
vertices[32*n+25] = uint8(x1 >> 8)
vertices[32*n+26] = uint8(y1)
vertices[32*n+27] = uint8(y1 >> 8)
vertices[32*n+28] = uint8(u1)
vertices[32*n+29] = uint8(u1 >> 8)
vertices[32*n+30] = uint8(v1)
vertices[32*n+31] = uint8(v1 >> 8)
} else { } else {
vertices = append(vertices, vertices[32*n] = uint8(x0 >> 8)
uint8(x0>>8), uint8(x0), uint8(y0>>8), uint8(y0), vertices[32*n+1] = uint8(x0)
uint8(u0>>8), uint8(u0), uint8(v0>>8), uint8(v0), vertices[32*n+2] = uint8(y0 >> 8)
uint8(x1>>8), uint8(x1), uint8(y0>>8), uint8(y0), vertices[32*n+3] = uint8(y0)
uint8(u1>>8), uint8(u1), uint8(v0>>8), uint8(v0), vertices[32*n+4] = uint8(u0 >> 8)
uint8(x0>>8), uint8(x0), uint8(y1>>8), uint8(y1), vertices[32*n+5] = uint8(u0)
uint8(u0>>8), uint8(u0), uint8(v1>>8), uint8(v1), vertices[32*n+6] = uint8(v0 >> 8)
uint8(x1>>8), uint8(x1), uint8(y1>>8), uint8(y1), vertices[32*n+7] = uint8(v0)
uint8(u1>>8), uint8(u1), uint8(v1>>8), uint8(v1)) vertices[32*n+8] = uint8(x1 >> 8)
vertices[32*n+9] = uint8(x1)
vertices[32*n+10] = uint8(y0 >> 8)
vertices[32*n+11] = uint8(y0)
vertices[32*n+12] = uint8(u1 >> 8)
vertices[32*n+13] = uint8(u1)
vertices[32*n+14] = uint8(v0 >> 8)
vertices[32*n+15] = uint8(v0)
vertices[32*n+16] = uint8(x0 >> 8)
vertices[32*n+17] = uint8(x0)
vertices[32*n+18] = uint8(y1 >> 8)
vertices[32*n+19] = uint8(y1)
vertices[32*n+20] = uint8(u0 >> 8)
vertices[32*n+21] = uint8(u0)
vertices[32*n+22] = uint8(v1 >> 8)
vertices[32*n+23] = uint8(v1)
vertices[32*n+24] = uint8(x1 >> 8)
vertices[32*n+25] = uint8(x1)
vertices[32*n+26] = uint8(y1 >> 8)
vertices[32*n+27] = uint8(y1)
vertices[32*n+28] = uint8(u1 >> 8)
vertices[32*n+29] = uint8(u1)
vertices[32*n+30] = uint8(v1 >> 8)
vertices[32*n+31] = uint8(v1)
} }
n++
} }
return vertices return vertices[:n*32]
} }